Applicant withdrawal behavior is of considerable interest to organizations and selection system designers. Some of the primary reasons for this interest are that applicant decisions to withdraw from a selection procedure can impact the size and quality of the applicant pool (Barber & Roehling, 1993), which can decrease the utility of the selection procedure (Murphy, 1986) and increase the potential for adverse impact (Ryan, Sacco, McFarland, & Kriska, 2000; Tam, Murphy, & Lyall, 2004). The current study builds a model of applicant withdrawal based on prior theoretical and empirical work and subsequently tests components of this model. The proposed predictors of withdrawal intentions and behavior include applicant perceptions, motivation, selection process features, employment background characteristics, and individual differences.

Data were collected from ∼ 25,000 applicants to a large manufacturing start-up company in the US at four time points in the selection process—application, post-test, post-assessment, and post-interview. Results from linear and logistic regression analyses provided support for the proposed model of applicant withdrawal, highlighting the importance of both applicant perceptions and contextual features.

More specifically, the study found that perceptions of Person-Job (P-J) and Person-Organization (P-O) fit as well as perceptions of offer expectancy were among the strongest predictors of withdrawal intentions. Withdrawal behavior was more difficult to predict, but was significantly predicted by current employment status, demographic characteristics, number of previous jobs held, P-O fit perceptions, and withdrawal intentions.

Implications of the study's results for organizations and selection system designers are discussed within the limitations of the current research. Additionally, directions for future research are detailed within a broader framework of applicant withdrawal.